Back

Unbounded Cell Growth and Proteome Imbalance in Doxorubicin-Induced Senescent RPE-1 Cells

Liu, X.; Sonnett, M.; Kirschner, M. W.

2025-10-28 cell biology
10.1101/2025.10.27.684969 bioRxiv
Show abstract

Cellular senescence is traditionally viewed as a terminal state of cell-cycle arrest accompanied by widespread molecular remodeling, yet its underlying regulatory logic and progression remain poorly understood. Here, we combined quantitative phase microscopy and normalized Raman imaging with quantitative proteomic and phosphoproteomic profiling to examine human RPE1 cells undergoing doxorubicin-induced senescence. Senescent cells did not reach a steady state but instead exhibited sustained, unbounded growth over a 12 day period, marked by a continuous rise in dry mass and volume coupled with declining mass density. Time resolved proteomics revealed extensive and asynchronous remodeling across organelles, with lysosomal, ER, and Golgi proteins increasing in abundance, whereas nuclear and mitochondrial proteins declined, indicating large scale reorganization of cellular composition. Phosphoproteomic inference linked these structural shifts to regulatory signaling, confirming the expected downregulation of CDK activity while revealing coordinated activation of stress and DNA damage responsive kinases such as CAMK2D, DNAPK, and MARK family members. Together, these integrated data depict senescence as a dynamic, actively regulated state, maintained through coordinated remodeling of proteome composition and signaling activity rather than passive arrest. Our findings highlight how combining quantitative biophysical measurements with multi-layered molecular profiling exposes the regulatory architecture that sustains the senescent phenotype and its loss of internal homeostasis.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.